Ceramide-Activated Protein Kinase

Proc. Natl. Acad. Sci. USA Vol. 93, pp. 6959-6963, July 1996 Biochemistry

Ceramide-binding and activation defines protein kinase c-Raf as a ceramide-activated protein kinase ANDREA HUWILER*t, JOSEF BRUNNERt, RICHARD HUMMEL*, MARGRIET VERVOORDELDONKt, SILVIA STABEL§, HENK VAN DEN BOSCHt, AND JOSEF PFEILSCHIFTER*¶ *Department of Pharmacology, Biozentrum, University of Basel, Klingelbergstasse 70, CH-4056 Basel, Switzerland; tCentre for Biomembranes and Lipid Enzymology, University of Utrecht, Padualaan 8, 3584 CH Utrecht, The Netherlands; tDepartment of Biochemistry, Swiss Federal Institute of Technology Zurich (ETHZ), ETH-Zentrum, CH-8092 Zurich, Switzerland; and §Max-Delbruck-Labor in der Max-Planck-Gesellschaft, Carl-von-Linne-Weg 10, D-50829 Koln, Germany Communicated by Tadeus Reichstein, Institut fiir Organische Chemie der Universitat Basel, Basel, Switzerland, April 5, 1996 (received for review February 12, 1996)

ABSTRACT Interleukin 1 is the prototype of an inflam- induce the expression of a specific type IV collagenase, a group matory cytokine, and evidence suggests that it uses the II phospholipase A2, eicosanoids, an inducible nitric oxide sphingomyelin pathway and ceramide production to trigger synthase, and a variety of chemokines in mesangial cells (11). mitogen-activated protein kinase (MAPK) activation and We have used rat mesangial cells as a model system to evaluate subsequent gene expression required for acute inflammatory IL-1-induced signaling events and report that IL-1 stimulates processes. To identify downstream signaling targets of cer- ceramide production, which subsequently specifically binds to amide, a radioiodinated photoaffinity labeling analog of cer- and activates protein kinase c-Raf. amide ([1251]3-trifluoromethyl-3-(m-iodophenyl)diazirine- ceramide) was employed. It is observed that ceramide specifically binds to and activates protein kinase c-Raf, leading to MATERIALS AND METHODS a subsequent activation of the MAPK cascade. Ceramide does Cell Culture. Rat renal mesangial cells were cultured as not bind to any other member of the MAPK module nor does it described (12). Single cells were cloned by limited dilution on bind to protein kinase C-c. These data identify protein kinase 96-microwell plates. Clones with apparent mesangial cell mor- c-Rafas a specific molecular target for interleukin 1,8-stimulated phology were used for further processing. The cells were grown ceramide formation and demonstrate that ceramide is a lipid in RPMI 1640 medium supplemented with 10% (vol/vol) fetal cofactor participating in regulation of c-Raf activity. calf serum, penicillin (100 units/ml), streptomycin (100 Ag/ ml), and bovine insulin (0.66 units/ml). For the reported Interleukin 1 (IL-1) is a major product of activated monocytes experiments, passages 5-18 of mesangial cells were used. and is also released by many cell types when exposed to an Lipid Extraction and Separation. Confluent cells in 30-mm inflammatory environment. The biological activities of IL-1 diameter dishes were labeled for 24 h with [14C]serine (0.3 are initiated by binding to two types of IL-1 receptors, desig- ,tCi/ml; specific activity, 53 mCi/mmol) and stimulated as nated IL-1 receptors type I and type II. Although the cyto- indicated. The reaction was stopped by extraction of lipids plasmic portions of both IL-1 receptors do not contain kinase (13), and sphingomyelin and ceramide were resolved by se- domains or motifs homologous to any other known signaling TLC pathway, rapid intracellular protein phosphorylation occurs in quential one-dimensional using chloroform/methanol/ response to IL-1 stimulation. The nature of the primary signal ammonia (65:35:7.5; vol/vol) followed by chloroform/ delivered by IL-1 receptor activation is poorly understood and methanol/acetic acid (9:1:1, vol/vol). Spots corresponding to controversial (1), but it is clear that IL-1 does stimulate protein ceramide and sphingomyelin were analyzed and quantitated using kinase activities in a wide variety of different cell types. IL-1 a Berthold (Nashua, NH) LB 2842 automatic TLC scanner. and tumor necrosis factor a have been shown to induce rapid In Vivo Ps-Labeling of Cells and Immunoprecipitation. mitogen-activated protein kinase (MAPK) activation in fibro- Confluent mesangial cells in 100-mm diameter dishes were blasts, U937 cells, KB cells, and mesangial cells (2-6). Fur- incubated for 2 days in DMEM containing 0.1 mg of fatty thermore, both cytokines employ the sphingomyelin signaling acid-free BSA per ml and then washed three times with pathway to generate ceramide and to stimulate a putative phosphate-free DMEM to remove all phosphate. Then the ceramide-activated serine/threonine protein kinase (7-9). cells were metabolically labeled for 4 h at 37°C with Lipid second messengers are increasingly recognized as im- [32P]orthophosphate (0.5 mCi/plate) in phosphate-free portant mediators of extracellular signals, and ceramide gen- DMEM. After labeling, cells were stimulated as indicated. To erated by the action of neutral and acidic sphingomyelinases stop the reaction, cells were washed twice with ice-cold buffer has been implicated in a variety of physiological processes, like containing 50 mM Tris HCl (pH 7.5) and 150 mM NaCl and effects on cell growth, differentiation, and apoptosis (8, 9). scraped into 1 ml of buffer A (50 mM Tris HCl, pH 7.5/150 Rat renal mesangial cells are a well-defined IL-i-responsive mM NaCl/10% glycerol/1% Triton X-100/2 mM EDTA/2 cell type that is involved in most pathological processes of the mM EGTA/40-mM ,B-glycerophosphate/50 mM NaF/10 mM renal glomerulus (10, 11). Resting mesangial cells do not sodium pyrophosphate/25 ,ug of leupeptin per ml/200 units of produce any inflammatory mediator constitutively but require aprotinin per ml/1 ,tM pepstatin A/1 mM phenylmethylsul- a triggering by invading immune cells. Three prominent fea- fonyl fluoride) and homogenized by 10 passes through a tures of mesangial cells evolve as a result of the crosstalk with 26-gauge needle. The homogenate was centrifuged for 10 min invading neutrophils and macrophages, as follows: increased at 14,000 x g, and the supernatant was taken for immunopre- mediator production, increased matrix synthesis, and increased cipitation. mesangial cell proliferation (10, 11). IL-1 has been reported to Abbreviations: IL-1, interleukin 1; MAPK, mitogen-activated protein The publication costs of this article were defrayed in part by page charge kinase; TID, 3-trifluoromethyl-3-(m-iodophenyl)diazirine; MEK, payment. This article must therefore be hereby marked "advertisement" in MAPK kinase; PKC, protein kinase C. accordance with 18 U.S.C. §1734 solely to indicate this fact. ITo whom reprint requests should be addressed. 6959 Downloaded by guest on September 28, 2021 6960 Biochemistry: Huwiler et al. Proc. Natl. Acad. Sci. USA 93 (1996) Samples (1 ml), containing 150 x 106 cpm of labeled X-100, 1 ,uM ATP, 1 ,uCi of [,y-32P]ATP, and 100 ng of purified proteins, 5% fetal calf serum, and 1.5 mM iodoacetamide in recombinant kinase-negative MEK (MEKk-). The reaction buffer A were incubated with the indicated antibodies for 3 h was stopped by addition of SDS/sample buffer and subjected at room temperature, or overnight at 4°C. Then 100 ,l of a to SDS/PAGE. After fixation in 25% isopropanol/10% acetic 50% (vol/vol) slurry of protein A-Sepharose 4B-CL in PBS acid, the gels were dried and exposed to Hyperfilm MP or was added, and the mixture was incubated for 1 h under mild analyzed with a PhosphorIm4ger (Molecular Dynamics). shaking. After centrifugation for 5 min at 3000 x g, immuno- Chemicals. [14C]Serine, [ly-32P]ATP, and [32P]orthophos- complexes were washed three times with 1 ml of a low-salt phate were from Amersham, C6- and C16-ceramide were from buffer (50 mM Tris HCl, pH 7.5/150 mM NaCl/0.2% Triton Calbiochem, IL-1f3 was provided by CIBA-Geigy, and protein X-100/2 mM EDTA/2 mM EGTA/0.1% SDS) and three A-Sepharose 4B-CL was from Pharmacia. The photoaffinity times with a high-salt buffer (50 mM Tris HCl, pH 7.5/500 mM labeling reagent [125I]TID-ceramide was synthesized as de- NaCl/0.2% Triton X-100/2 mM EDTA/2 mM EGTA/0.1% scribed (17). SDS). Pellets wer boiled for S min in 50 gl of Laemmli dissociation buffer and subjected to SDS/PAGE. 3-Trifluoromethyl-3-(m-iodophenyl)diazirine (TID) Label- A PY MAPK PKCt MEK c-Raf MEKK ing Studies. Confluent mesangial cells were incubated for 2 days in DMEM containing 0.1 mg of fatty acid-free BSA per (kDa) - + - + - +-_ + - + - +1 ml. Thereafter cells were stimulated for 5 min with 50 nM N-[3-[[[2-(125I)Iodo-4-[3-(trifluoromethyl)-3H-diazirin-3-yl]- 210- benzyl]oxy]carbonyl]propanoyl]-D-erythro-sphingosine ([125I]TID-ceramide; 0.2 mCi/ml) and subjected to photolysis 97- for 30 s using a high-pressure mercury lamp (Osram HBO, 350 W; Osram, Berlin, Germany) mounted in a SUSS LH 1000 lamphouse equipped with a shutter to control exposure time. 68- Cells were then lysed in buffer A homogenized by 10 passes through a 26-gauge needle fitted to a 1-ml syringe and centrifuged for 10 min at 14,000 x g, and the supernatant was taken for immunoprecipitation and subjected to SDS/PAGE. 45- The antibodies for c-Raf, p42 and p44 isoforms of MAPK, MAPK kinase (MEK), and protein kinase C (PKC)-; have been characterized in detail elsewhere (14-16). After fixation in 25% isopropanol/10% acetic acid, the gels were dried and exposed to Hyperfilm MP (Amersham) at -70°C. 29- MAPK Activity Assay. MAPK activity assays were per- formed as described (15). In brief, stimulated cells were lysed and homogenized in kinase buffer (20 mM Tris HCl, pH 7.5/1 B mM EGTA/2 mM MnCl2/0.1 mM sodium orthovanadate/25 (kDa) 1 2 3 4 5 6 ,tg of leupeptin per ml/200 units of aprotinin per ml/1 ,uM 210 pepstatin A/1 mM phenylmethylsulfonyl fluoride) by 10 passes through a 26-gauge needle, centrifuged for 5 min at 14,000 x g, and the supernatant was taken for determination of protein 97- concentration. Cell extracts (50 ,tg) were incubated for 15 min at 30°C in the presence of 20 pLg of myelin basic protein, 10 ,uM 68 - ATP, and 2 ,uCi of [y-32P]ATP. The reaction was terminated by adding Laemmli buffer, and samples were subjected to SDS/PAGE (14% acrylamide gel) followed by autoradiogra- phy to visualize the phosphorylation of myelin basic protein. 45 - c-RafActivity Assay. Stimulated cells were lysed in buffer A and homogenized by 10 passes through a 26-gauge needle. Homogenates were centrifuged for 10 min at 14,000 x g, and the supernatant was taken for immunoprecipitation of c-Raf 29 - using a C terminus-specific polyclonal antibody as described (14). Immunoprecipitates were incubated for 10 min at 37°C in a 50-gl kinase reaction containing 10 mM Tris HCl (pH 7.5), FIG. 1. Specific binding of [125I]TID-ceramide to c-Raf in vivo. (A) 10 mM MnCl2, 150 mM NaCl, 2 mM dithiothreitol, 1% Triton Mesangial cells were stimulated for 5 min with [1251]TID-ceramide then directly lysed (-) or photolyzed before lysis (+), and equal Table 1. Stimulation of ceramide production and sphingomyelin amounts of protein were taken from immunoprecipitation using hydrolysis by IL-1l3 specific antibodies against phosphotyrosine (PY; at a dilution of 1:200), p42maPk -and p44mapk (MAPK; 1:100 each), PKC-; (1:100), Ceramide, Sphingomyelin, MEK (1:100), c-Raf (1:250), and MEK kinase (1:100). Immunopre- Addition % of control % of control cipitates were separated on SDS/PAGE and exposed to Hyperfilm MP Control 100 ± 5 100 ± 2 at -70°C. (B) Exogenous ceramide and IL-1,B pretreatment of mes- ± ± angial cells decrease [1251]TID-ceramide binding to c-Raf. Mesangial IL-1,8 536 25* 67 9** cells were stimulated for 5 min with 5OnM [125I]TID-ceramide (0.2 Mesangial cells were labeled for 24 h with [14C]serine and stimulated mCi/ml) and directly lysed (lane 2) or photolyzed before lysis (lanes for 10 min with 1 nM IL-1,B. Lipids were extracted and separated on 1 and 3-6). Binding of [125I]TID-ceramide was inhibited by addition TLC as described. Spots corresponding to ceramide and sphingomy- of 450 nM (lane 4) or 50 nM (lane 5) exogenous unlabeled ceramide elin were analyzed and quantitated. Results are expressed as percent- or by pretreatment of the cells with lnM of IL-1lB for 2 min before age of the respective control values and are means ± SD (n = 3). TID-ceramide addition (lane 6).Thereafter c-Raf was immunoprecipi- Significant differences from corresponding control: *, P < 0.001; ** tated (lanes 2-6; lane 1, preimmune serum), subjected to SDS/PAGE, P < 0.01 (by ANOVA). and exposed to Hyperfilm MP at -70°C. Downloaded by guest on September 28, 2021 Biochemistry: Huwiler et al. Proc. Natl. Acad. Sci. USA 93 (1996) 6961 A c-Raf activity (% of control)

-e- C-6 --C-16

C-1 6 (jiM) Co 1 10 100

.... 4._ 4. . _

10 100 1000 Ceramide (gM) B c-Raf activity (% of control) C-1 6 (gM) 700 r.I Co 1 10 100 600oo C-6 -A-C-16

500 -

400 -

300!-

200 -

1001

0.1 1 10 100 1000 Ceramide (,uM)

C C6-Cer C16 -Cer (AM)

Co 1 10 100 Co 1 10 30 100

Sit.'t:_ '4 M B P

FIG. 2. (A) Stimulation of c-Raf activity by C6-ceramide and C16-ceramide in vitro. c-Raf was immunoprecipitated from quiescent mesangial cells. Immunoprecipitates were incubated in vitro for 15 min at 30°C with the indicated concentrations of C6 and C16 ceramide before the addition of 100 ng of recombinant MEKk- and 1 ,uCi of [-y-32P]ATP for 10 min. Samples were subjected to SDS/PAGE, exposed to Hyperfilm MP, and quantitated on a Phosphorlmager. Data are expressed as percentage of the respective control values and are means of three independent experiments (the SD ranges from 12 to 34%). (Inset) Autoradiograph of one representative experiment. (B) Stimulation of c-Raf activity by C6-ceramide and C16-ceramide in vivo. Quiescent intact mesangial cells were stimulated with the indicated concentrations C6- and C16-ceramide for 5 min. Thereafter cell lysates were prepared and c-Raf was immunoprecipitated. Immunoprecipitates were incubated for 10 min at 30°C with 100 ng of recombinant MEKk- and 1 ,uCi of [-y-32P]ATP. Samples were subjected to SDS/PAGE, exposed to Hyperfilm MP, and quantitated on a Phosphorlmager. Data are expressed as percentage of the respective control values and are means of two independent experiments giving similar results. (Inset) Autoradiograph of one representative experiment. (C) Effect of C6- and C16-ceramides on MAPK activity in mesangial cells. Quiescent mesangial cells were stimulated with the indicated concentrations C6- and C16-ceramide for 5 min. Thereafter myelin basic protein phosphorylation was measured as described.

RESULTS mesangial cells, we prepared a photoaffinity labeling analog of ceramide of high [12 I]iodine-specific radioactivity (>2000 Upon exposure to IL-1fB, mesangial cells hydrolyze sphingo- Ci/mmol) (17). Incubation of intact cells with ['25I]TID- myelin to generate ceramide (Table 1) and display a rapid ceramide for 5 min followed by homogenization and sequential activation of MAPK, which is sustained for >24 h (5). To immunoprecipitation of the different members of the MAPK identify possible molecular targets of ceramide signaling in module with specific polyclonal antibodies reveals a selective Downloaded by guest on September 28, 2021 6962 Biochemistry: Huwiler et al. Proc. Natl. Acad. Sci. USA 93 (1996) labeling of protein kinase c-Raf (Fig. 1A), which is increased DISCUSSION up to 243 ± 37% (mean + SD, n = 4) upon UV-induced The serine/threonine kinase c-Raf is a well-studied signaling labeling. There is no labeling of the p42 and p44 isoforms of molecule that is responsible for phosphorylation and activation MAPK nor of their upstream activator MEK, MEK kinase, of MEK in the classical MAPK cascade (19). The mechanism PKC-C, or any protein immunoprecipitated with an anti- of activation of protein kinase c-Raf has been studied exten- phosphotyrosine antibody (PY) as shown in Fig. 1A. All sively, and it is now clear that.this is a multistep event. In a first antibodies used in this study have been shown to immunopre- step, c-Raf translocates to the plasma membrane and associ- cipitate their respective antigens (14-16). Labeling by ates with Ras-GTP. However, this association with Ras-GTP [1251]TID-ceramide is inhibited by exogenous addition of in- is not sufficient for c-Raf activation but is required for its creasing concentrations of unlabeled ceramide (Fig. 1B) as recruitment to the plasma membrane (20,21). In a second step, well as by endogenous ceramide produced by IL-113 prestimu- c-Raf is activated by an unknown mechanism that may com- lation of the cells (Fig. IB). These data establish that IL-1,8- prise tyrosine and/or serine/threonine phosphorylation of induced ceramide specifically binds to c-Raf in mesangial cells. c-Raf (14, 19, 22-25) or the interaction with another mem- Comparable data were obtained for tumor necrosis factor brane cofactor such as a lipid (18, 24, 25). The c-Raf amino a-stimulated mesangial cells (data not shown). terminus contains a highly. conserved region (CR-1) that Next we investigated whether ceramide binding to c-Raf also encompasses a zinc-finger motif analogous to the lipid-binding has functional consequences. Fig. 2A shows that C6 and C16 domain of PKC (19), and it is tempting to speculate that c-Raf analogs of ceramide, in a concentration-dependent manner, is activated by binding of a lipid second messenger in away similar increase protein kinase c-Raf activity in an immunocomplex to the mechanism of activation of PKC by 1,2-diacylglycerol. kinase assay in vitro. Moreover, C6 and C16 analogs also The present data clearly demonstrate that ceramide specif- activate protein kinase c-Raf in vivo when added to intact cells ically binds to c-Raf and stimulates its kinase activity. These and subsequent evaluation of c-Raf activity by an immuno- observations extend previous reports on putative lipid factors complex kinase assay (Fig. 2B). Fig. 2C demonstrates that extractable with chloroform/methanol from membranes of ceramide analogs not only activate c-Raf but that the signal is activated cells that markedly enhance c-Raf enzymatic activity (19, 25) and identify ceramide as a lipid second messenger that further processed along the MAPK cascade and causes an acts as a major direct activator of c-Raf in IL-1l3-triggered increased activity of the p42 and p44 isoforms of MAPK. signal propagation. Ceramide has gained recognition as an Ceramide-stimulated activation of MAPK is inhibited by PD important signaling molecule regulating fundamental biolog- 098059 (18), a synthetic inhibitor of MEK (data not shown). To ical processes, like cell proliferation and differentiation, on- prove that IL-1l3 indeed uses this signaling pathway to activate cogenesis, and immune and inflammatory processes (8, 9). c-Raf in mesangial cells, we examined c-Raf phosphorylation Ceramide is produced by sphingomyelin hydrolysis by both and activity. As shown in Fig. 3, IL-1,B induces a rapid neutral and acid sphingomyelinases. With respect to immedi- phosphorylation and activation of c-Raf within 5 min and ate targets of ceramide, suggestions for several ceramide- subsequent stimulation of MAPK (5). activated enzymes have been forwarded. These include a 97-kDa proline-directed serine/threonine protein kinase (26, A IL-13 27), a ceramide-activated protein phosphatase (28) and PKC-G co 1' 5' 10' 30' 90' (29). Recently Yao et al. (30) observed that a 97-kDa ceram- kD- 97- ide-activated protein kinase phosphorylates c-Raf on Thr-269 -7 c-Raf_ and increases its activity toward MEK. Furthermore, in intact HL-60 cells, ceramide-activated protein kinase forms complexes with c-Raf and, in response to tumor necrosis factor a, 45- phosphorylates and activates c-Raf. In IL-1,3-stimulated mesangial cells, ceramide definitely does not bind to PKC-C and there is also no ceramide-binding protein in the range of 97 B kDa visible in the cell homogenates (data not shown). In c-Raf activity (% of control) 160- contrast, our data define c-Raf as a ceramide-activated protein "I kinase, thus suggesting that there is more than one ceramide- activated protein kinase and that ceramide may have cell-type 140 specific targets. Previous reports suggested that short-chain ceramide ana- 120 - logs are biologically more active than their long-chain equiv- alents likely due to the increased solubility of the former. Based on this background, the observation that the C16- 100 -- ceramides were more potent the C6-ceramides in activating c-Raf is rather unexpected. Moreover, we observed that 80- stimulation of c-Raf activity in vivo occurs at lower concentrations of both ceramide derivatives than in vitro. 60 A possible explanation could be that C16-ceramide was a i ~~~ii natural product, whereas C6-ceramide was a synthetic com- pound. In addition to chain length and solubility in aqueous 40 -- .__ Control IL-lB solution, other factors may be important for biological activity, especially the degree of unsaturation of the alkyl chains. FIG. 3. IL-l,B-induced phosphorylation and activation of c-Raf in In summary, IL-113 signaling in mesangial cells involves vivo. (A) Quiescent mesangial cells were labeled with 32Pi and were ceramide generation and subsequent direct binding to and stimulated with IL-1l3 for the indicated time periods. c-Raf was Two important aspects of this study are the immunoprecipitated, subjected to SDS/PAGE, and exposed to Hy- activation of c-Raf. perfilm MP. (B) Cells were treated with vehicle (control) or IL-lb (1 identification of ceramide as the long-missing lipid activator of nM) for 5 min, and c-Raf activity in mesangial cell homogenates was c-Raf and the identification of protein kinase c-Raf as another, determined as described. Results are expressed as percentage of yet molecularly defined, member of an emerging family of control and are means ± SD of three independent experiments. ceramide-activated protein kinases. Activation of protein ki- Downloaded by guest on September 28, 2021 Biochemistry: Huwiler et al. Proc. Natl. Acad. Sci. USA 93 (1996) 6963 nase c-Raf by ceramide is probably only one possible way to 13. Bligh, E. C. & Dyer, W. J. (1995) Can. J. Biochem. Physiol. 37, activate this kinase, especially in response to cytokines like 911-917. IL-1f3 or tumor necrosis factor a. Activation of c-Raf by 14. Sozeri, O., Vollmer, K., Liyanage, M., Frith, D., Kour, G., Mark, G. E., III, & Stabel, S. (1992) Oncogene 7, 2259-2262. phosphorylation is an alternative mechanism used by tyrosine 15. Huwiler, A., Stabel, S. Fabbro, D. & Pfeilschifter, J. (1995) kinase receptor agonists or G-protein-coupled receptors, Biochem. J. 305, 777-784. which do not trigger sphingomyelin hydrolysis. 16. Huwiler, A., Fabbro, D., Stabel, S. & Pfeilschifter, J. (1992) FEBS Some pressing questions that immediately arise are con- Lett. 300, 259-262. cerned with the molecular mechanism of ceramide binding to 17. Weber, T. & Brunner, J. (1995) J. Am. Chem. Soc. 117, 3084- and activation of c-Raf and the possibility that ceramide 3095. stimulation of c-Raf also signals to other MAPK modules like 18. Dudley, D. T., Pang, L., Decker, S. J., Bridges, A. J. & Saltiel, the stress-activated protein kinases (31), as has recently been A. R. (1995) Proc. Natl. Acad. Sci. USA 92, 7686-7689. 19. Daum, G., Eisenmann-Tappe, I., Fries, H.-W., Toppmair, J. & suggested for HL-60 cells (32). Rapp, U. R. (1994) Trends Biochem. Sci. 19, 474-480. 20. Leevers, S. J., Paterson, H. F. & Marshall, C. J. (1994) Nature This work was supported by an European Molecular Biology (London) 369, 411-414. Laboratory shortterm fellowship to A.H., by Swiss National Science 21. Stokoe, D., Macdonald, S. G., Cadwallader, K., Symons, M. & Foundation Grants 31-43090.95 to J.P. and 31-36193.92 to J.B., by Hancock, J. F. (1994) Science 264, 1463-1467. grants from the Commission of the European Union (Biomed 2, PL 22. Kolch, W., Heidecker, G., Kochs, G., Hummel, R., Vahidi, H., 950 979), by a grant from the Wilhelm Sander-Stiftung, and by grants Mischak, H., Finkenzeller, G., Marme, D. & Rapp. U. R. (1993) of the Roche Research Foundation and the CIBA-Geigy Jubilaums- Nature (London) 364, 249-252. stiftung. 23. Belka, C., Wiegmann, K., Adam, D., Holland, R., Neuloh, M., Herrmann, F., Kr6nke, M. & Brach, M. A. (1995) EMBO J. 14, 1. Dinarello, C. A. (1991) Blood 77, 1627-1652. 1156-1165. 2. Guy, G. R., Cluna, S. P., Wong, N. S., Ng, S. B. & Tan, Y. H. 24. Marais, R., Light, Y., Paterson, H. F. & Marshall, C. J. (1995) (1991) J. Bio. Chem. 266, 14343-14352. EMBO J. 14, 3136-3145. 3. Bird, T. A., Sleath, P. R., de Ross, P. C., Dower, S. K. & Virca, 25. Dent, P., Reardon, D. B. Morrison, D. K. & Sturgill, T. W. (1995) G. D. (1991) J. Biol. Chem. 266, 22661-22670. Mol. Cell. Biol. 15, 4125-4135. 4. Guesdon, F., Freshney, N., Waller, R. J., Rawlinson, L. & Sak- 26. Mathias, S., Dressler, K. A. & Kolesnick, R. N. (1991) Proc. Nat!. latvala, J. (1993) J. Biol. Chem. 268, 4236-4243. Acad. Sci. USA 88, 10009-10013. 5. Huwiler, A. & Pfeilschifter, J. (1994) FEBS Lett. 350, 135-138. 27. Joseph, C. K., Byun, H.-S., Bittman, R. & Kolesnick, R. N. (1993) 6. Wiegmann, K., Schutze, S., Machleidt, T., Witte, D. & Kronke, J. Biol. Chem. 268, 20002-20006. M. (1994) Cell 78, 1005-1015. 28. Dobrowsky, R. T., Kamibayashi, C., Mumby, M. C. & Hannun, 7. Mathias, S., Younes, A., Kan, C., Orlow, I., Joseph, C. & Y. A. (1993) J. Biol. Chem. 268, 15523-15530. Kolesnick, R. N. (1993) Science 259, 519-522. 29. Muller, G., Ayoub, M., Storz, P., Rennecke, J., Fabbro, D. & 8. Kolesnick, R. & Golde, D. W. (1994) Cell 77, 325-328. Pfizenmaier, K. (1995) EMBO J. 14, 1961-1969. 9. Hannun, Y. A. & Obeid, L. M. (1995) Trends Biochem. Sci. 20, 30. Yao, B., Zhang, Y., Delikat, S., Mathias, S., Basu, S. & Kolesnick, 73-77. R. (1995) Nature (London) 378, 307-310. 10. Kashgarian, M. & Sterzel, R. B. (1992) Kidney Int. 41, 524-529. 31. Cano, E. & Mahadevan, L. C. (1995) Trends Biochem. Sci. 20, 11. Pfeilschifter, J. (1994) News Physiol. Sci. 9, 271-276. 117-122. 12. Pfeilschifter, J., & Vosbeck, K. (1991) Biochem. Biophys. Res. 32. Westwick, J. K., Bielawska, A. E., Dbaibo, G., Hannun, Y. A. & Commun. 175, 372-379. Brenner, D. A. (1995) J. Biol. Chem. 270, 22679-22692. Downloaded by guest on September 28, 2021